A groundwater regime detection method and system based on water flow numerical model simulation

Abstract

The application discloses a groundwater condition detection method and system based on water flow numerical model simulation, and relates to data processing technology, and comprises the following steps: a topographic model containing a target region is established; a proportional relationship in the topographic model is determined, and a section is performed on multiple target directions of the topographic model based on the proportional relationship; a plurality of representative sites are determined according to the section, stratum data of the representative sites is collected, and the collected stratum data is mapped to corresponding sections; fitting is performed on the corresponding sections based on the mapped stratum of the representative sites, so as to construct stratum distribution; a data sequence is constructed according to obtained correlation monitoring data of groundwater of a plurality of collection points and rainfall data of the target region; and groundwater condition data of each section of an isolated island topographic model is completed by using a preset completion network model according to the data sequence. The groundwater condition detection is realized in combination with a stratum structure model and a completion network mode, and a new method for detecting the groundwater condition is provided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a groundwater condition detection method and system based on water flow numerical model simulation. BACKGROUND

[0002] Groundwater has the characteristics of stable water quantity and good water quality, and is one of the important water sources for agricultural irrigation, enterprise production and urban resident life. In particular, in arid and semi-arid areas where surface water is lacking, groundwater is often used as the main water source. Groundwater resources play a crucial role in regional water supply problems. If groundwater resources are not properly managed, overuse of groundwater resources can cause problems such as land subsidence and collapse, leading to local water resource depletion and even groundwater pollution, causing serious environmental problems.

[0003] Traditional groundwater condition estimation is based on various hydraulic tests and numerical models to calculate the available water quantity of the groundwater system. Modeling is time-consuming and requires a large number of hydrogeological parameters. Due to factors such as complex geological conditions, uncertainty of water condition prediction parameters, and mathematical modeling burden, a relatively accurate regional groundwater condition quantitative prediction tool has not yet been formed, and it is not possible to refine the accurate prediction of regional groundwater conditions in each part, and it cannot serve the regional water supply management well. SUMMARY

[0004] The embodiments of the present application provide a groundwater condition detection method and system based on water flow numerical model simulation, which realizes groundwater condition detection by combining stratum structure models and a way of completing network architecture design, and proposes a new method for detecting groundwater conditions.

[0005] The embodiments of the present application propose a groundwater condition detection method based on water flow numerical model simulation, comprising:

[0006] Obtain topographic elevation data and establish a topographic model containing a target region;

[0007] Determine the scale relationship in the topographic model, and perform sectioning in multiple target directions of the topographic model based on the scale relationship to obtain an isolated island topographic model of the target region, wherein the sectioning depth at least completely covers the groundwater region;

[0008] Determine a plurality of representative locations according to the sectioning, and collect stratum data of the representative locations, and map the collected stratum data to the corresponding sectioning;

[0009] Fit the stratum of the mapped representative locations in the corresponding sectioning to construct stratum distribution in each sectioning of the isolated island topographic model, and determine the initial groundwater distribution for each sectioning of the isolated island topographic model;

[0010] constructing a data sequence according to the obtained correlation monitoring data of the groundwater of the plurality of collection points and rainfall data of the target region;

[0011] completing the groundwater condition detection according to the data sequence by using a preset completion network model to complete the groundwater condition data of each section of the isolated island topographic model.

[0012] Optionally, after obtaining the isolated island topographic model of the target region, the method further comprises:

[0013] determining a stratum permeability coefficient distribution according to the collected stratum data of the representative location;

[0014] adjusting the section depth of each section of the isolated island topographic model based on the stratum permeability coefficient distribution, and

[0015] extracting a geological feature orientation based on the stratum data, and preferentially cutting along the direction of the fault zone according to the geological feature orientation.

[0016] Optionally, the cutting in the plurality of target directions of the topographic model based on the proportional relationship comprises: setting the target directions of the plurality of sections such that the sections in the plurality of target directions are perpendicular to the horizontal plane, and the sections are parallel or intersected with each other.

[0017] fitting in the corresponding sections based on the mapped stratum of the representative location to construct the stratum distribution in each section of the isolated island topographic model comprises:

[0018] fitting in the current section according to the adjusted section depth of the section and the extracted geological feature orientation; and

[0019] correcting the fitting result according to the geological features of the associated sections parallel and perpendicular to the current section.

[0020] Optionally, the correcting the fitting result according to the geological features of the associated sections parallel and perpendicular to the current section comprises:

[0021] extracting stratum mutation features of the associated sections parallel and perpendicular to the current section;

[0022] searching for similar stratum patterns along the parallel section direction based on the stratum mutation features;

[0023] constructing a Bayesian probability model and a spatial constraint in the perpendicular section direction to calculate the stratum extension, wherein the spatial constraint is used to make the adjacent sections meet the continuity at the junction.

[0024] correcting the fitting result according to the distribution of the calculation results of the stratum extension based on each associated section.

[0025] Optionally, the modifying the fitting result according to the geological features of the associated cross sections parallel and perpendicular to the current cross section further comprises modifying the karst cave structure by the following steps:

[0026] For the current cross section, judging the permeability coefficient mutation feature according to the collected stratum data representing the site; and,

[0027] Identifying similar features of the permeability coefficient mutation feature at the associated positions in the at least two adjacent parallel or intersecting associated cross sections;

[0028] Determining the karst cave structure according to the identified similar features.

[0029] Optionally, the preset completion network model comprises a spatio-temporal graph convolution network (ST-GCN) and a temporal convolution network (TCN).

[0030] The completing the groundwater regime data of each cross section of the isolated island topography model according to the data sequence by using the preset completion network model comprises:

[0031] Determining the geographical positions of each collection point and calculating the three-dimensional Euclidean distances between the collection points;

[0032] Setting a dynamic adjacency matrix based on the calculated three-dimensional Euclidean distances;

[0033] Constructing a spatial topology graph according to the dynamic adjacency matrix, wherein the associated monitoring data of the groundwater of the collection points are taken as the node features of the spatial topology graph;

[0034] Extracting spatio-temporal graph features of the spatial topology graph by using a spatio-temporal graph convolution (ST-GCN) to complete the data based on the spatio-temporal graph features.

[0035] Optionally, the completing the groundwater regime data of each cross section of the isolated island topography model according to the data sequence by using the completion network model further comprises:

[0036] Extracting rainfall features of the target region by using a temporal convolution network (TCN);

[0037] Projecting the extracted rainfall features and the spatio-temporal graph features to a unified dimension:

[0038] Fusing the rainfall features and the spatio-temporal graph features in the unified dimension to complete the groundwater regime data based on the fused features.

[0039] Optionally, the completing the groundwater regime data based on the fused features is implemented by using a decoder.

[0040] The embodiment of the present application also provides a groundwater condition detection system based on water flow numerical model simulation, comprising a processor and a memory, wherein the memory stores a computer program, and the computer program is executed by the processor to realize the steps of the groundwater condition detection method based on water flow numerical model simulation.

[0041] The embodiment of the present application provides a new groundwater condition detection method, which can detect the groundwater conditions of multiple sections of an island terrain model by establishing the island terrain model and complementing the groundwater condition data of each section of the island terrain model based on a complement network model, and better serves the groundwater management of urban, town and rural areas.

[0042] The above description is only a summary of the technical scheme of the present application, in order to more clearly understand the technical means of the present application, and can be implemented according to the content of the specification, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are intended to depict only preferred embodiments of the application, and therefore should not be considered to narrow the scope of the present application in any way. Instead, they are included to provide illustration of the embodiments of the application in conformity with the principles of the present disclosure. In the drawings:

[0044] Figure 1 The basic flow of the groundwater condition detection method of the embodiment is shown in the figure;

[0045] Figure 2 The data complementing flow of the groundwater condition detection method of the embodiment based on a complement network model is shown in the figure. DETAILED DESCRIPTION

[0046] Exemplary embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings. While example embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood, and the scope of the present disclosure can be accurately conveyed to those skilled in the art.

[0047] The embodiment of the present application provides a groundwater condition detection method based on water flow numerical model simulation, as shown in the figure, comprising the following steps: Figure 1

[0048] ​In step S101, terrain elevation data is acquired, and a terrain model containing a target region is established. For example, a terrain model containing a target region can be established according to terrain elevation data DEM, and the target region is a region in which groundwater condition detection needs to be performed.

[0049] In step S102, a scale relationship in the terrain model is determined, and a section is performed in multiple target directions of the terrain model based on the scale relationship, to obtain an isolated island terrain model of the target region, wherein a section depth at least completely covers a groundwater region. That is, a section is performed in multiple directions based on the established terrain model, and stratum data after the section is filled according to subsequent acquisition data. In this way, a stratum structure from the ground surface inward is established, which provides a basis for effective analysis and display of groundwater conditions.

[0050] In step S103, multiple representative locations are determined according to the section, and stratum data of the representative locations is acquired, and the acquired stratum data is mapped to corresponding sections. In a specific example, an existing monitoring well or the like can be used as a representative location, or an acquisition point can be added, so as to acquire stratum data of the representative location. The representative locations can be distributed, and one representative location can exist in multiple sections.

[0051] In step S104, the stratum of the mapped representative locations is fitted in the corresponding sections, to construct a stratum distribution in each section of the isolated island terrain model, and to determine an initial groundwater distribution for each section of the isolated island terrain model. Through fitting, a stratum distribution of the target region can be obtained, and an initial groundwater distribution of the isolated island terrain model is determined according to a historically proven underground condition.

[0052] In step S105, a data sequence is constructed according to acquired correlation monitoring data of groundwater of multiple acquisition points and rainfall data of the target region. In a specific example, the correlation monitoring data of groundwater can include a permeability coefficient, a stratum water storage rate, lithology, and the like, and also includes a monitoring water level of the acquisition point. In a specific example, the groundwater condition also has a strong correlation with regional rainfall, and therefore, the embodiment of the present application further constructs a data sequence based on rainfall data.

[0053] In step S106, a preset completion network model is used to complete groundwater condition data of each section of the isolated island terrain model according to the data sequence, to complete groundwater condition detection. After completion, a groundwater condition distribution in multiple sections of the isolated island terrain model can be obtained, so as to complete detection.

[0054] The embodiment of the present application proposes a new method for detecting groundwater conditions. By establishing an island terrain model and combining a completion network model to complete the groundwater condition data of each section of the island terrain model, the groundwater conditions of multiple sections under the island terrain model can be detected, which better serves the groundwater management of urban, town and rural areas.

[0055] In some embodiments, after obtaining the island terrain model of the target area, the method further comprises:

[0056] According to the stratum data of the representative location, a stratum permeability coefficient distribution is determined.

[0057] Based on the stratum permeability coefficient distribution, the section depth of each section of the island terrain model is adjusted. By adjusting the section depth, the trend of the groundwater conditions can be represented in the island terrain model.

[0058] Based on the stratum data, a geological feature guide is extracted, and the section is preferentially cut along the fracture zone direction according to the geological feature guide. In a specific example, the geological feature guide represents the distribution and trend of different strata under the ground surface, and preferentially cutting the section along the fracture zone direction according to the geological feature guide can improve the accuracy of capturing the karst cave structure in karst terrain areas.

[0059] In some embodiments, cutting the section in multiple target directions of the terrain model based on the proportional relationship comprises: setting the target directions of the multiple sections such that the sections of the multiple target directions are perpendicular to the horizontal plane, and the sections are parallel or intersected with each other. That is, each section is perpendicular to the horizontal plane, and the sections can be parallel or intersected with each other.

[0060] Based on the mapping of the stratum of the representative location in the corresponding section, fitting is performed to construct the stratum distribution in each section of the island terrain model.

[0061] According to the section depth of the adjusted section and the extracted geological feature guide, fitting is performed in the current section. In a specific example, three-dimensional fitting can be performed according to the distribution of the geological feature guide of the discrete collection points on the multiple sections and the adjusted section depth to obtain the stratum condition under the island terrain model.

[0062] The fitting result is corrected according to the geological features of the associated sections parallel and perpendicular to the current section.

[0063] In some embodiments, correcting the fitting result according to the geological features of the associated sections parallel and perpendicular to the current section comprises:

[0064] The stratum mutation features of the associated sections parallel and perpendicular to the current section are extracted, such as obvious changes of rock stratum and soil stratum.

[0065] Based on the formation mutation feature, similar formation patterns are searched along the parallel section direction. For example, formations with similar permeability coefficients, lithology, and other parameters can be searched along the parallel section direction according to the data of each acquisition point.

[0066] A Bayesian probability model is constructed in the vertical section direction, and a spatial constraint is used to calculate the formation extension, in which the spatial constraint is used to make adjacent sections meet the continuity at the intersection. In a specific example, the Bayesian probability model is constructed in the vertical section direction based on the acquisition points, and the specific spatial constraint can be determined according to the geological feature orientation and the feature continuity at the intersection of the sections.

[0067] According to the distribution of the formation extension calculation results based on the associated sections of each associated section, the fitting results are corrected. After correction, different formations can be presented by different colors or structures in the isolated island terrain model and each section.

[0068] In some embodiments, the fitting results are corrected according to the geological features of the associated sections parallel and perpendicular to the current section also include correcting the cave structure by the following steps:

[0069] For the current section, according to the collected formation data representing the site, the permeability coefficient mutation feature is judged. In a specific example, if there is continuous formation distribution, the permeability coefficient will not exist in the case of mutation, and the possible cave structure is detected by judging the permeability coefficient mutation.

[0070] In the associated positions of at least two adjacent parallel or intersecting associated sections, the similar features of the permeability coefficient mutation feature are identified, and the similar features of the mutation are further identified by identifying the possible cave coverage positions in the associated sections.

[0071] The size and range of the cave structure are determined according to the identified similar features.

[0072] In some embodiments, the preset completion network model includes a space-time graph convolution network ST-GCN and a time convolution network TCN. The embodiment of the application designs a completion network model architecture, which is implemented based on a space-time graph convolution network ST-GCN and a time convolution network TCN, as shown in Figure 2 .

[0073] According to the data sequence, the underground water emotion data of each section of the isolated island terrain model is completed by using the preset completion network model, which includes:

[0074] The geographical positions of each acquisition point are determined, and the three-dimensional Euclidean distances between the acquisition points are calculated.

[0075] A dynamic adjacency matrix is set based on the calculated three-dimensional Euclidean distances. The adjacency matrix A ij can be defined as when dij ≤ time, A ij = exp (-d² ij / σ²) , otherwise 0. Here the parameters s and R are determined by seepage test, such as s = 500 meters, R = 2 kilometers.

[0076] According to the dynamic adjacency matrix, a spatial topology graph is constructed, wherein the correlation monitoring data of the groundwater of the collection points are taken as node features of the spatial topology graph, such as taking the correlation monitoring data of the permeability coefficient, the formation water storage rate, the water level change, and the like as the node features.

[0077] The spatio-temporal graph features of the spatial topology graph are extracted by using a spatio-temporal graph convolution ST-GCN, so as to perform data completion based on the spatio-temporal graph features. Specifically, the graph convolution operation can be performed based on the spatial topology graph.

[0078] In some embodiments, as shown in FIG. 1, the data sequence is utilized to complete the groundwater condition data of each section of the isolated terrain model by using a completion network model, which further includes: Figure 2

[0079] The rainfall data of the target region are utilized to extract rainfall features by using a time convolution network TCN, for example, a dilated causal convolution stack can be adopted, the dilation factors are 1, 2, 4, and 8, the convolution kernel size is set to 5, and each layer contains a residual connection.

[0080] The extracted rainfall features and the spatio-temporal graph features are projected to a unified dimension, and the output Hg of the ST-GCN and the output Ht of the TCN can be projected to the unified dimension.

[0081] The rainfall features and the spatio-temporal graph features in the unified dimension are fused, so as to complete the groundwater condition data based on the fused features. In a specific example, the fused features can be generated by calculating the correlation of the spatio-temporal features. Then, the decoder is utilized to complete the groundwater condition data based on the fused features. Through the scheme of the present application, the efficiency and the accuracy of the groundwater detection can be improved based on the ST-GCN and the TCN, and the result of the water condition detection is presented by using the isolated terrain model, thereby providing a new efficient auxiliary scheme for the groundwater management.

[0082] The present application also provides a groundwater condition detection system based on the water flow numerical model simulation, which includes a processor and a memory, the memory stores a computer program, and the computer program is executed by the processor to realize the steps of the groundwater condition detection method based on the water flow numerical model simulation.

[0083] ​Furthermore, although example embodiments have been described herein, the scope of coverage of embodiments includes any and all embodiments having equivalent elements, modifications, omissions, combinations (e.g., of

[0084] The above description is intended to be illustrative, and not restrictive. For example, the above-described examples (or one or more aspects thereof) can be used in combination with each other. Other embodiments will be readily apparent to those skilled in the art from the description herein, in conjunction with its drawings.

[0085] The above embodiments are only exemplary embodiments of the present disclosure, and those skilled in the art can make various modifications or equivalent replacements to the present disclosure within the spirit and protection scope of the present disclosure, and such modifications or equivalent replacements should also be considered to fall within the protection scope of the present disclosure.

Claims

1. A method for detecting a groundwater condition based on a numerical simulation of a water flow, characterized by, The method comprises: acquiring terrain elevation data and establishing a terrain model containing a target region; determining a scale relationship in the terrain model and performing sectioning in multiple target directions of the terrain model based on the scale relationship to obtain an isolated island terrain model of the target region, wherein the sectioning depth at least completely covers a groundwater region; determining multiple representative sites according to the sectioning and collecting stratum data of the representative sites, and mapping the collected stratum data to corresponding sections; fitting the mapped stratum of the representative sites in the corresponding sections to construct stratum distribution in each section of the isolated island terrain model, and determining an initial groundwater distribution for each section of the isolated island terrain model; constructing a data sequence according to acquired correlation monitoring data of groundwater of multiple collection points and rainfall data of the target region; completing groundwater detection by using a preset completion network model to complete groundwater data of each section of the isolated island terrain model according to the data sequence; after obtaining the isolated island terrain model of the target region, the method further comprises: determining stratum permeability coefficient distribution according to the collected stratum data of the representative sites; adjusting the sectioning depth of each section of the isolated island terrain model based on the stratum permeability coefficient distribution, and extracting a geological feature guide based on the stratum data and preferentially sectioning along a fault zone direction according to the geological feature guide; the sectioning in multiple target directions of the terrain model based on the scale relationship comprises: setting the target directions of multiple sections such that the sections in multiple target directions are perpendicular to a horizontal plane, and the sections are parallel or intersected with each other; the fitting of the mapped stratum of the representative sites in the corresponding sections to construct stratum distribution in each section of the isolated island terrain model comprises: fitting in a current section according to the adjusted sectioning depth of the section and the extracted geological feature guide; and correcting the fitting result according to geological features of associated sections parallel and perpendicular to the current section.

2. The groundwater condition detecting method based on water flow numerical model simulation according to claim 1, wherein, the correcting the fitting result according to the geological features of the associated sections parallel and perpendicular to the current section comprises: extracting stratum mutation features of the associated sections parallel and perpendicular to the current section; searching for similar stratum patterns along a parallel section direction based on the stratum mutation features; constructing a Bayesian probability model and a spatial constraint in a perpendicular section direction to calculate stratum extension, wherein the spatial constraint is used to make adjacent sections meet continuity at a boundary; correcting the fitting result according to a distribution of the stratum extension calculation results based on the associated sections.

3. The groundwater condition detecting method based on water flow numerical model simulation according to claim 1, wherein, the correcting the fitting result according to the geological features of the associated sections parallel and perpendicular to the current section further comprises correcting a karst cave structure by the following steps: judging permeability coefficient mutation features according to the collected stratum data of the representative sites for the current section; and identifying similar features of the permeability coefficient mutation features at associated positions in at least two adjacent associated sections that are parallel or intersected with each other; determining a karst cave structure according to the identified similar features.

4. The groundwater condition detecting method based on water flow numerical model simulation according to claim 1, wherein, the preset completion network model comprises a spatio-temporal graph convolution network ST-GCN and a temporal convolution network TCN; The step of completing the groundwater condition data of each section of the isolated terrain model according to the data sequence by using the preset completion network model comprises: determining the geographical positions of each collection point and calculating the three-dimensional Euclidean distances between the collection points; setting a dynamic adjacency matrix based on the calculated three-dimensional Euclidean distances; constructing a spatial topology graph based on the dynamic adjacency matrix, wherein the associated monitoring data of the groundwater of the collection points are taken as the node features of the spatial topology graph; extracting the spatio-temporal graph features of the spatial topology graph by using a spatio-temporal graph convolution ST-GCN, and completing the data based on the spatio-temporal graph features.

5. The groundwater condition detecting method based on water flow numerical model simulation according to claim 4, wherein, The step of completing the groundwater condition data of each section of the isolated terrain model according to the data sequence by using the completion network model further comprises: extracting rainfall features of the rainfall data of the target region by using a time convolution network TCN; projecting the extracted rainfall features and the spatio-temporal graph features to a unified dimension; fusing the rainfall features and the spatio-temporal graph features in the unified dimension, and completing the groundwater condition data based on the fused features.

6. The groundwater condition detecting method based on water flow numerical model simulation according to claim 4, wherein, The step of completing the groundwater condition data based on the fused features is implemented by using a decoder.

7. A groundwater condition detection system based on numerical simulation of water flow, characterized by, The device comprises a processor and a memory, and the memory stores a computer program which, when executed by the processor, implements the steps of the groundwater condition detection method based on the simulation of the water flow numerical model according to any one of claims 1 to 6.

Images